Best thermal grease?

courtney01

Gawd
Joined
Nov 25, 2008
Messages
619
I tried searching for what the best thermal grease is out there, but I can only find results from at least half a year ago, and so I wanted to just make sure things haven't changed since then and ask what the best I can get today is. I only need enough for 2 computers since I'm not building any more than that and I'm only a noob, not a build enthusiast. Something affordable but good.
 
They're all withen delta 2C of each other, so I wouldnt worry too much about it. That said, Thermalright's stuff comes in a nice big tube with lots of it, so you're good for multiple applications, the old standby: AS5, comes in this tiny little package thats good for 1 or 2 applications.
 
They're all withen delta 2C of each other, so I wouldnt worry too much about it. That said, Thermalright's stuff comes in a nice big tube with lots of it, so you're good for multiple applications, the old standby: AS5, comes in this tiny little package thats good for 1 or 2 applications.

I got about 10-15 builds out of a 3.5g tube of AS5. In retrospect you get 4g's of paste with a Thermalright cooler. MX-2 and TX-2 also come in 4g tubes.

Like said before, they are all within 2-3c of each other. I found AS5 easier to work with than MX-2 or TX-2, but the AS5 needs a good 200 hours to cure where the others don't require cure times.

Matter of personal preference really. Buy a tube of each. It's cheap. Experiment for yourself.
 
I use Shin-etsu X23 however the AC MX-2 is supposedly very good. When I run out of Shin-etsu Ill get some MX-2.
 
the thermalright stuff, I just bought a new TRUE for my i7, it is considered good? I didnt realize it was up there with the likes of AS5.
 
Like said before, they are all within 2-3c of each other. I found AS5 easier to work with than MX-2 or TX-2, but the AS5 needs a good 200 hours to cure where the others don't require cure times

What does curing mean? And is it better or worse if a material doesn't need to cure? (I don't mean better or worse in terms of convenience of not having to wait, but rather in terms of materials being thermally better or worse)

What do you mean by "easier to work" with?
 
Curing time is the time is takes for the TIM to reach optimal thermal conductivity. AS5 requires roughly 200 hours to cure properly. MX-2 and TX-2 do not require curing, they are good to go from the start.

Easier to work with meaning easier to spread. Not as viscous. I found MX-2 to be kinda stiff, or lumpy compared to AS5.
 
Curing time is the time is takes for the TIM to reach optimal thermal conductivity. AS5 requires roughly 200 hours to cure properly. MX-2 and TX-2 do not require curing, they are good to go from the start.

Easier to work with meaning easier to spread. Not as viscous. I found MX-2 to be kinda stiff, or lumpy compared to AS5.

Oh I see. So not requiring any curing doesn't somehow mean lesser thermal conductivity than one that does require it, right?
 
Oh I see. So not requiring any curing doesn't somehow mean lesser thermal conductivity than one that does require it, right?

Nope, the only difference with a paste that requires curing is that it'll take a while before it gets to its peak performance level.
 
Do I need to wait out the entire 200 hours before I can turn on the computer, or can I start it up before?
 
You need it to be on and running for 200 hours before it gives you the best results :D

Oh cool! I was sad that there was going to be such a long wait time before I get to use my computer. I'll just get the AS5 then! Thanks
 
Do I need to wait out the entire 200 hours before I can turn on the computer, or can I start it up before?

The curing period is during operation, meaning that those 200 hours are while the computer is running. So, like the above poster said, the computer actually has to be on for that period, or rather, the burn-in period only occurs when the computer is on.
 
The curing period is during operation, meaning that those 200 hours are while the computer is running. So, like the above poster said, the computer actually has to be on for that period, or rather, the burn-in period only occurs when the computer is on.

Almost identical posting time as my previous one! I'm glad I got this clarified, or else I would have been going crazy waiting for over a week.

Do you recommend leaving my computer on continuously for that amount of time, or just use it normally?
 
Just use it normally, unless of course normal for you is leaving it on all the time (it is for me).
 
the MX2 is generally 1C better than the cured AS5, and it is still very easy to work with.

OCZ freeze and Tim Consultants grease 0098 might be slightly better than MX2, but all the other compounds at the top are difficult to work with (IC7 Diamond / Cool Labs Liquid pro) and I'm not convinced that they perform any better.

Edit/Update: http://benchmarkreviews.com/index.p...sk=view&id=138&Itemid=62&limit=1&limitstart=3

You need to consider some margin of error for their testing, but that should be a good summary. I suppose there they show MX2 on par with AS5 (Usually I see a slight nod to MX2).
 
You need it to be on and running for 200 hours before it gives you the best results :D

i think you have to turn it on and off a couple times... at least that's what it says at the bottom of this page:

http://www.arcticsilver.com/as5.htm

but yeah, "normal usage" should be more than enough.

as for which one is better, i honestly don't think it matters that much. applying the compound correctly is probably more important than the actual compound.

i wish i could find that article with instructions on the best way to apply thermal compound based on the shape of the heatsink base. i thought it was a very useful guide.
 
I personally like IC Diamond 7, @ $10/tube w/ tax and at my micro center, there's nothing that beats it. Better than AS5 and cheaper than everything else but at the same price point.
 
Anybody have experience with (or helpful knowledge of) OCZ freeze or TIM Consultants T-C Grease 0098?

I was initially just going to go with the AS5, but after reading the link archmage referred me to, I might just go for something without a cure time. The two above are at least as good as AS5 right? Are they used as widely as AS5?
 
I personally like IC Diamond 7, @ $10/tube w/ tax and at my micro center, there's nothing that beats it. Better than AS5 and cheaper than everything else but at the same price point.

How easy is it to work with? In what ways would it be better than AS5?
 
AS5, comes in this tiny little package thats good for 1 or 2 applications.

Geeze, how much are you using?! I've had my AS5 for about 4 years and still have plenty left. I've built about 7 computers since then while using this stuff on graphics cards and reseating heatsinks after lapping. One small tube (they do sell a larger tube too) will last quite some time!
 
Yea that's just ridiculous... AS5 tube will last a long time, as will tubes of almost any thermal paste.

Right now my processor is on IC7 Diamond, and I found it a PITA to install to be honest. It's not worth the effort. I re-mounted it several times to test temperatures, and the only time I actually got great results was when I pre-heated everything (heatsink included) with a heat gun. This stuff just won't spread easily.

I have used MX-2. It's easy. It worked well on the 1st try with no fuss.

Obviously I've used AS5 (and 1,2, and 3). They're all easy.

About the Tim Consultants 0098: I've never actually used it (or even seen it). However I've read that there's nothing difficult about it. Jab-tech (no experience with them, just an example) has it for $5, and I'm sure some other places have it by now because it has gained some popularity.

Application method: Even though these sites are testing application methods, it is my personal experience that it doesn't seem to matter much as long as I "feel" that each area is getting adequate and level contact.

End note: It really doesn't matter for a personal build. Don't wait an extra week because you want 1C better temps. It probably wasn't worth that week of waiting. The curing time on AS5 is nothing to worry about.
 
I'm the ultimate noob. :( I just learned from somebody that intel CPUs comes with thermal grease already applied on there. But I'm glad I learned about all these. My system isn't going to be overclocked nor is it a hardcore gaming rig. How much better are these products, like AS5, than the stock material? Do you recommend using non-stock thermal compounds?
 
For the money? TX-2. Performs as well as MX-2 or AS5 (better, really) at 1/2 the price. Non conducting, too.

Google some comparison tests.
 
I'm the ultimate noob. :( I just learned from somebody that intel CPUs comes with thermal grease already applied on there. But I'm glad I learned about all these. My system isn't going to be overclocked nor is it a hardcore gaming rig. How much better are these products, like AS5, than the stock material? Do you recommend using non-stock thermal compounds?

I would NEVER use the stock compound, as it is oftentimes wrongly applied and unreliable. Some other guy used the stock paste on his processor, and it idling so high the computer would shut down. He ended up using some ArcticSilver5 and problem solved. Always buy paste.
 
I would NEVER use the stock compound, as it is oftentimes wrongly applied and unreliable. Some other guy used the stock paste on his processor, and it idling so high the computer would shut down. He ended up using some ArcticSilver5 and problem solved. Always buy paste.

Do I need to remove the stock compound before applying my own? If so, should I just use isopropyl alcohol? I read somewhere that's what people use.

For the money? TX-2. Performs as well as MX-2 or AS5 (better, really) at 1/2 the price. Non conducting, too.

Google some comparison tests.

Is AS5 non-conducting too?
 
Do I need to remove the stock compound before applying my own? If so, should I just use isopropyl alcohol? I read somewhere that's what people use.



Is AS5 non-conducting too?

It is non conducting, however it is capacitive, meaning that if you have 2 paths of electricity flow, it could bridge them if very close together.
 
It is non conducting, however it is capacitive, meaning that if you have 2 paths of electricity flow, it could bridge them if very close together.

Is tx-2 also capacitive? Does tx-2 have a long enough track record to use reliably over the long run? I read a review of tx-2 made 1 year ago that says tx-2 is just as good or better than as5, but the only thing to be wary about is that it hasn't been used as long as as5, so it's not known whether it'll be reliable in the long run like the as5. will there be degradation or cracking in the tx-2 after several years? I'm a bit worried now about AS5 because someone warned me that it could affect the electronics on the board if any gets on them.
 
Anybody have experience with (or helpful knowledge of) OCZ freeze or TIM Consultants T-C Grease 0098?

I was initially just going to go with the AS5, but after reading the link archmage referred me to, I might just go for something without a cure time. The two above are at least as good as AS5 right? Are they used as widely as AS5?


I've used AS5 and OCZ Freeze, and both have been about the same temperature wise for me. I have noticed that it is easier to separate the heatsink from the cpu with OCZ than with AS5 when I have switched cpus or changed heatsinks.
 
Is tx-2 also capacitive? Does tx-2 have a long enough track record to use reliably over the long run? I read a review of tx-2 made 1 year ago that says tx-2 is just as good or better than as5, but the only thing to be wary about is that it hasn't been used as long as as5, so it's not known whether it'll be reliable in the long run like the as5. will there be degradation or cracking in the tx-2 after several years? I'm a bit worried now about AS5 because someone warned me that it could affect the electronics on the board if any gets on them.

All thermal grease will eventually break down. Realistically, most people replace their tech before the 3-5 year window during which you might need to replace your thermal compound.

That said, yes, TX-2 and MX-2 both last longer than AS5, last I heard.

You'll find hard-liners on either side, but I generally go with what's best. Imo, AS5 is simply an outdated formula.
 
As of right now, I'm testing my new motherboard to see if it works, so I'm not putting anything together yet. I haven't received my RAM or video card, but I do have my CPU, so I went out to my local store and bought the cheapest RAM and video card just to use for this mobo test. For the purpose of this test only, not for my permanent build, is it ok to just use the stock thermal grease already applied to the CPU, then after testing it, remove the stock paste and reapply a better one when I'm actually putting it all together with my final components?
 
I read I can use isopropyl alcohol to remove thermal paste (as close to 100% concentration as possible). What are possible wiping materials I can use?

Are the cloths that come with prescription glasses lint-free cloth?

When I first set my CPU in the socket, I tried to wipe off my finger prints from its back side (the side you apply the thermal paste to) using toilet paper, but it didn't work. Was using toilet paper a bad idea?
 
I read I can use isopropyl alcohol to remove thermal paste (as close to 100% concentration as possible). What are possible wiping materials I can use?

Are the cloths that come with prescription glasses lint-free cloth?

When I first set my CPU in the socket, I tried to wipe off my finger prints from its back side (the side you apply the thermal paste to) using toilet paper, but it didn't work. Was using toilet paper a bad idea?

I usually just use kleenex or paper towel, but it really doesn't matter that much. As long as you wet it first, the paper won't leave any lint behind. And there was nothing wrong with you using toilet paper, but you can't just wipe off finger prints. You need to use some sort of cleaning agent to remove it first, since the oils from your skin will get stuck in the pores of the metal heatspreader. Isopropyl alcohol will do the trick. 70% or higher is generally a good rule of thumb, although I've used 50% without any problems.
 
Any suggestions to properly applying AS5 onto an i7 920? I've always heard use a pea-sized dot and then squash it with the CPU heatsink and allow it to bake over time. I've also seen some people manually spread it over, which should I go with?
 
Back
Top